- crank everything! (didn't save numbers)
- added 4 rlu layers with around 1000 neurons
- very non-deterministic performance
- didn't understand model
- read the docs!
- larger input provides more info
- makes it harder to learn relevant parts
- may require longer training times
lanesSide = 0;
patchesAhead = 1;
patchesBehind = 0;
trainIterations = 10000;
temporal_window = 3;
hidden layers
layer_defs.push({
type: 'fc',
num_neurons: 1,
activation: 'relu'
});
looked up activation functions a relu seems best for beginner case
lanesSide = 6;
patchesAhead = 10;
patchesBehind = 10;
trainIterations = 10000;
- still just one neuron
- doesn't change lanes
- given one nueron is probably just stop/go?
layer_defs.push({
type: 'fc',
num_neurons: 100,
activation: 'relu'
});
- changes lanes inferquently
- still just rides behind one car and passes
avg speed: 55.47mph
- not very good
- probably undertrained
- probably don't need to look so far behind
- want to experiment with more layers
- more and less neurons
trainIterations = 50000;
- taking quite some time to train
- avg speed: 55.04 mph
- extra training didn't help
lanesSide = 6;
patchesAhead = 20;
patchesBehind = 8;
- huge input! 1471
- training is slow
- not really sure what the red graph is
- the boxes below appear to be the neural network, but so far only useful for understanding the input size
- starting to get impatient on the training...
- still not changing lanes
- going to reload page
lanesSide = 6;
patchesAhead = 20;
patchesBehind = 8;
trainIterations = 20000;
layer_defs.push({
type: 'fc',
num_neurons: 10,
activation: 'relu'
});
- not great at changing lanes
- maxes speed in open road
avg speed: 54.94
10 neurons did as well as 100 !
lanesSide = 6;
patchesAhead = 20;
patchesBehind = 8;
trainIterations = 20000;
layer_defs.push({
type: 'fc',
num_neurons: 500,
activation: 'relu'
});
layer_defs.push({
type: 'fc',
num_neurons: 500,
activation: 'relu'
});
made the net way too big let's trim this all down a bit
lanesSide = 5;
patchesAhead = 10;
patchesBehind = 2;
trainIterations = 10000;
layer_defs.push({
type: 'fc',
num_neurons: 10,
activation: 'relu'
});
- doing just as well as the big net!
- still getting stuck
- wondering what the appropriate tradeoffs are
avg speed: 56.6
- now seeing a bit of lane changin
- remember i have to be patient for traning to happen
- bigger isn't always better
- esp b.c training is compute AND memory bound
lanesSide = 5;
patchesAhead = 15;
patchesBehind = 2;
trainIterations = 20000;
763 bits
layer_defs.push({
type: 'fc',
num_neurons: 10,
activation: 'relu'
});
layer_defs.push({
type: 'fc',
num_neurons: 10,
activation: 'relu'
});
avg speed: 51.01 mph!!
lanesSide = 5;
patchesAhead = 5;
patchesBehind = 2;
trainIterations = 10000;
layer_defs.push({
type: 'fc',
num_neurons: 8,
activation: 'relu'
});
layer_defs.push({
type: 'fc',
num_neurons: 4,
activation: 'relu'
});
avg speed: 58.4
lanesSide = 3;
patchesAhead = 8;
patchesBehind = 2;
trainIterations = 10000;
layer_defs.push({
type: 'fc',
num_neurons: 8,
activation: 'relu'
});
layer_defs.push({
type: 'fc',
num_neurons: 4,
activation: 'relu'
});
avg speed: 51.48
lanesSide = 3;
patchesAhead = 8;
patchesBehind = 2;
trainIterations = 10000;
layer_defs.push({
type: 'fc',
num_neurons: 8,
activation: 'relu'
});
layer_defs.push({
type: 'fc',
num_neurons: 4,
activation: 'relu'
});
avg speed: 51.48
lanesSide = 4;
patchesAhead = 10;
patchesBehind = 3;
trainIterations = 20000;
layer_defs.push({
type: 'fc',
num_neurons: 32,
activation: 'relu'
});
layer_defs.push({
type: 'fc',
num_neurons: 16,
activation: 'relu'
});
avg speed: 69.87!!!
lanesSide = 3;
patchesAhead = 10;
patchesBehind = 2;
trainIterations = 10000;
layer_defs.push({
type: 'fc',
num_neurons: 32,
activation: 'relu'
});
layer_defs.push({
type: 'fc',
num_neurons: 16,
activation: 'relu'
});
lanesSide = 3;
patchesAhead = 10;
patchesBehind = 2;
trainIterations = 10000;
layer_defs.push({
type: 'fc',
num_neurons: 64,
activation: 'relu'
});
layer_defs.push({
type: 'fc',
num_neurons: 64,
activation: 'relu'
});
lanesSide = 3;
patchesAhead = 10;
patchesBehind = 2;
trainIterations = 10000;
layer_defs.push({
type: 'fc',
num_neurons: 64,
activation: 'relu'
});
layer_defs.push({
type: 'fc',
num_neurons: 64,
activation: 'relu'
});
avg: 69.39